AI Types of Learning
Choose a learning setup by the feedback actually available.
How it works
Supervised learning uses target labels, unsupervised learning looks for structure without those labels, and reinforcement learning uses rewards from actions in an environment. These are useful distinctions rather than an exhaustive taxonomy. The code describes three different kitchen-study records; it does not train three models.
Run the code
Try in playground loads this complete standard-library Python example without executing it. Choose Run when ready. Local review uses Python 3.12.14; the existing browser editor uses Pyodide 0.27.7. These small fictional examples need no account, external data, model-provider key or database service.
Example
records = [
{"task": "predict minutes", "feedback": "label"},
{"task": "group queues", "feedback": "none"},
{"task": "choose route", "feedback": "reward"},
]
kind = {"label": "supervised", "none": "unsupervised", "reward": "reinforcement"}
for row in records:
print(row["task"] + ": " + kind[row["feedback"]])
Expected result
predict minutes: supervised group queues: unsupervised choose route: reinforcement
A missing label does not automatically make a dataset suitable for clustering. You still need meaningful features and a reason to compare records. Reward learning also needs a defined environment and an evaluation policy; a static route planner is not automatically a reinforcement learner.